Accurate and Scalable Classification of Colonoscopy Neoplasia using Machine Learning and Natural Language Processing

Abstract

Introduction: Colorectal cancer (CRC) remains a leading cause of cancer associated death in the United States and colonoscopy the primary screening strategy for prevention. Rates of adenomatous and serrated neoplasia detection are inversely associated with post-colonoscopy CRC. This crucial quality metric depends on accurate ascertainment of colorectal neoplasia findings from both endoscopy and histopathology records. We aimed to assess the feasibility of a random forest machine learning model to rapidly and accurately categorize colorectal neoplasia from electronic health record data. Methods: A retrospective cohort study compared neoplasia detection rates among individuals undergoing colonoscopy at a large academic institution to develop a rule-based algorithm to categorize colorectal neoplasia from endoscopy reports and pathology SNOMED II codes. This cohort provided a large training set to develop a natural language processing (NLP) system using a random forest approach to automatically classify unstructured pathology findings into adenoma, serrated, or advanced neoplasms. This system was manually validated through an independent holdout set. Results: The training set comprised 35,953 unstructured pathology reports with matched SNOWMED II codes from 95,188 unstructured colonoscopy reports. The final model was assessed on an independent holdout set of 337 manually annotated procedures obtaining an AUC of 0.997 (CI 0.994 - 1), 0.99 (CI 0.98-1), and 0.99 (CI 0.98-0.99) for prediction of adenoma, serrated, and advanced lesions respectively. Discussion: The random forest-based hybrid NLP system for classification of colonoscopy results was both accurate and explainable. NLP combined with effective machine learning algorithms can provide a scalable strategy for colonoscopy quality monitoring.

Affiliated Institutions

• Jacksonville College US
• Mayo Clinic US
• Health First US
• Mayo Clinic in Arizona US
• Mayo Clinic Health System US
• WinnMed US
• Mayo Clinic in Florida US

Related Publications